Download The twin arginine protein transport pathway exports multiple

Molecular Microbiology (2010) 77(1), 252–271 䊏
doi:10.1111/j.1365-2958.2010.07206.x
First published online 7 June 2010
The twin arginine protein transport pathway exports
multiple virulence proteins in the plant pathogen
Streptomyces scabies
mmi_7206 252..271
Madhumita V. Joshi,1† Stefan G. Mann,2†
Haike Antelmann,3 David A. Widdick,4,5
Joanna K. Fyans,4,6 Govind Chandra,2
Matthew I. Hutchings,5 Ian Toth,6 Michael Hecker,3
Rosemary Loria1 and Tracy Palmer4*
1
Department of Plant Pathology and Plant-Microbe
Biology, Cornell University, Ithaca, NY 14853, USA.
2
Department of Molecular Microbiology, John Innes
Centre, Norwich NR4 7UH, UK.
3
Institute for Microbiology, Ernst-Moritz-Arndt-University
of Greifswald, D-17487 Greifswald, Germany.
4
Division of Molecular Microbiology, College of Life
Sciences, University of Dundee, Dundee DD1 5EH, UK.
5
School of Biological Sciences, University of East
Anglia, Norwich Research Park, Norwich, NR4 7TJ, UK.
6
Scottish Crop Research Institute, Invergowrie, Dundee,
DD2 5DA, UK.
Summary
Streptomyces scabies is one of a group of organisms
that causes the economically important disease
potato scab. Analysis of the S. scabies genome
sequence indicates that it is likely to secrete many
proteins via the twin arginine protein transport (Tat)
pathway, including several proteins whose coding
sequences may have been acquired through horizontal gene transfer and share a common ancestor with
proteins in other plant pathogens. Inactivation of the
S. scabies Tat pathway resulted in pleiotropic phenotypes including slower growth rate and increased permeability of the cell envelope. Comparison of the
extracellular proteome of the wild type and DtatC
strains identified 73 predicted secretory proteins that
were present in reduced amounts in the tatC mutant
strain, and 47 Tat substrates were verified using a Tat
reporter assay. The DtatC strain was almost completely avirulent on Arabidopsis seedlings and was
delayed in attaching to the root tip relative to the
Accepted 4 May, 2010. *For correspondence. E-mail t.palmer@
dundee.ac.uk; Tel. (+44) (0)1382 386464; Fax (+44) (0)1382 388216.
†
These authors contributed equally to this work.
© 2010 Blackwell Publishing Ltd
wild-type strain. Genes encoding 14 candidate Tat
substrates were individually inactivated, and seven of
these mutants were reduced in virulence compared
with the wild-type strain. We conclude that the Tat
pathway secretes multiple proteins that are required
for full virulence.
Introduction
Potato scab is a polyphyletic disease caused by organisms in the genus Streptomyces, the best studied of which
is Streptomyces scabies (Loria et al., 2006). Like other
streptomycetes, S. scabies is a mycelial bacterium that
undergoes complex morphological differentiation involving the formation of aerial hyphae and spores (see Flärdh
and Buttner, 2009 for a review of the complex biology of
Streptomyces). The hyphal form of S. scabies infects
plants through expanding plant tissue, including roots and
tubers (Loria et al., 2006; 2008).
One of the major pathogenicity determinants of S.
scabies is thaxtomin, a nitrated dipeptide toxin that is a
potent inhibitor of cellulose synthesis. The toxin induces
plant cell hypertrophy in expanding plant tissues and likely
facilitates penetration of plant tissue by the pathogen (King
et al., 1989; Healy et al., 2000; Scheible et al., 2003; Bischoff et al., 2009). The mechanism by which thaxtomin
inhibits cellulose synthesis is undefined but it apparently
interacts with a highly conserved target, as the toxin affects
all higher plants; this is congruent with the fact that
thaxtomin-producing streptomycetes have an extremely
wide host range. The thaxtomin biosynthesis genes are
conserved in the plant pathogenic species, including
Streptomyces turgidiscabies and Streptomyces acidiscabies, and are regulated by TxtR, an AraC family transcriptional regulator that binds cellobiose as a co-inducer (Kers
et al., 2005; Joshi et al., 2007a). The S. scabies 87-22
genome contains a biosynthetic cluster that is highly
similar in structure and organization to the coronafacic
acid (CFA) clusters from the Gram negative plant pathogens Pseudomonas syringae, and Pectobacterium
atrosepticum (Bignell et al., 2010). CFA is the polyketide
component of coronatine that acts as a jasmonate mimic
during plant interactions; mutational studies demonstrate
The Tat pathway in S. scabies 253
that the cluster contributes to virulence in S. scabies 87-22
as it does in Gram negative pathogens. Unlike the thaxtomin biosynthetic cluster, the coronafacic acid-like cluster
is not conserved in S. turgidiscabies or S. acidiscabies.
Secreted proteins are also critical determinants in the
interaction between bacteria and eukaryotic hosts; virulence proteins are secreted either into the host environment or directly into host cells. Indeed, a secreted
proteinaceous virulence factor, Nec1, lacks homologues
outside of plant pathogenic streptomycetes, is required for
colonization of the roots and may function to suppress
plant defence responses (Bukhalid and Loria, 1997; Joshi
et al., 2007b). Furthermore, the secreted protein TomA, is
conserved among pathogenic streptomycetes (Kers et al.,
2005) and is homologous to products of saponinaseencoding genes, which are important for host–pathogen
interactions in some plant pathogenic fungi (Seipke and
Loria, 2008).
The role of protein secretion in pathogenesis is particularly well characterized in the case of Gram negative
bacteria, which have numerous systems dedicated to
achieving the secretion of proteins across the complex
double-membrane cell envelope (see, e.g. Christie et al.,
2005; Büttner and He, 2009; Galán, 2009). In contrast,
Gram positive bacteria have a simpler cell envelope, and
secretion generally requires passage of proteins across
only a single membrane. Therefore, it might be expected
that the general protein transport machineries residing in
the prokaryotic cytoplasmic membrane play more direct
roles in the virulence of Gram positive organisms. In
support of this it has been shown that many Gram positive
bacteria have an accessory SecA protein, termed SecA2
that appears to contribute to bacterial virulence (e.g. Rigel
and Braunstein, 2008).
The Tat system is, like Sec, a general protein export
pathway that is found in the cytoplasmic membranes of
some (although not all) bacteria and archaea. In Gram
negative bacteria and in the Gram positive Actinobacteria,
that include the streptomycetes, the Tat system is comprised of three essential components, TatA, TatB and TatC
(Bogsch et al., 1998; Sargent et al., 1998; 1999; Weiner
et al., 1998; Schaerlaekens et al., 2004; Hicks et al.,
2006). By contrast, the Tat machineries in the low G + C
Gram positive bacteria (exemplified by Bacillus subtilis)
and in the archaea do not require a TatB protein and the
Tat systems in these prokaryotes are made up only of
TatA and TatC components (Jongbloed et al., 2004; Dilks
et al., 2005).
Proteins are targeted to the Tat pathway by means of
N-terminal signal peptides that superficially resemble Sec
signal peptides, but that harbour a conserved S/T-R-R-xF-L-K consensus motif, where the twin arginines are
invariant and normally essential for efficient export by the
Tat pathway (Berks, 1996; Stanley et al., 2000). The main
distinguishing feature of the Tat system is that it transports
fully folded proteins across the cytoplasmic membrane. In
bacteria such as Escherichia coli and B. subtilis, relatively
few proteins are exported by the Tat pathway and, in E.
coli at least, the majority of them contain complex redox
cofactors that are assembled into the substrate protein
prior to transport across the membrane (reviewed in
Palmer et al., 2005; Lee et al., 2006).
Although in most organisms the Tat pathway is generally held to be a relatively minor route of protein export,
there is evidence that in some halophilic archaea, and in
particular in Streptomyces bacteria, Tat exports significant
numbers of proteins. Thus, bioinformatic predictions on
the genome sequence of Streptomyces coelicolor suggested that as many as 189 proteins may be substrates of
the Tat pathway (Rose et al., 2002; Dilks et al., 2003;
Bendtsen et al., 2005). The validity of these prediction
programmes were largely borne out by proteomic studies
of S. coelicolor wild type (WT) and tat mutant strains
coupled with testing of candidate Tat signal peptides in
Tat-dependent reporter assays, where a total of 33 Tat
substrate proteins have now been confirmed (Li et al.,
2005; Widdick et al., 2006).
In spite of the fact that the Tat system is a general
protein export system, Tat-secreted proteins contribute
to virulence in some Gram negative and Gram positive
bacteria (e.g. Ochsner et al., 2002; Ding and Christie,
2003; Pradel et al., 2003; Saint-Joanis et al., 2006).
Given the apparent importance of the Tat pathway in
protein secretion in the streptomycetes, we reasoned
that this pathway may play an essential role in the virulence of S. scabies. Analysis of the recently available S.
scabies 87-22 genome sequence (http://www.sanger.
ac.uk/Projects/S_scabies/) with Tat signal peptide prediction programmes suggests that in excess of 100
proteins may be exported by the Tat pathway in this
organism, including several that appear to share a
common ancestor with homologues in other plant pathogens. Through proteomic studies coupled with reporter
protein secretion assays we have verified 47 Tat substrates in this organism. Importantly, we show that a
DtatC mutant of S. scabies is almost completely avirulent, and that the Tat pathway secretes multiple proteins
that are required for full virulence.
Results
Bioinformatic analysis of the S. scabies genome
sequence indicates that it encodes many candidate
Tat substrates
Two generally available prediction programmes,
TATFIND 1.4 and TatP, have been developed to identify
candidate Tat-targeting signal peptides (Rose et al.,
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
254 M. V. Joshi et al. 䊏
2002; Bendtsen et al., 2005). When these programmes
are applied to all of the open reading frames (ORFs)
encoded by the genome sequence of S. scabies,
TATFIND 1.4 predicts 154 likely Tat substrates, while
TatP predicts 177 (the list of these proteins is available
at http://www.lifesci.dundee.ac.uk/groups/tracy_palmer/
links.html). However, both of these programmes generate a degree of false positive and false negative
predictions; for example, when applied to the genome
sequence of E. coli, both programmes overpredict Tat
substrates by 20–25% (Dilks et al., 2003; Bendtsen
et al., 2005). In general, the sets of candidate Tat substrates predicted by the two programmes show partial
but not complete overlap. However to date, where
tested, all of those predicted to be Tat substrates by both
TATFIND 1.4 and TatP have been shown to have bona
fide Tat-targeting signals (Widdick et al., 2006). This
therefore gives confidence that the subset of proteins
predicted to be Tat substrates by both programmes are
highly likely to represent real substrates. For S. scabies,
82 proteins are predicted to be Tat substrates by each of
the two programmes, and these are listed in Table S1.
It was noted from previous work with S. coelicolor that
it is sometimes difficult to identify the correct N-terminus
of a predicted protein and this can lead to mis-annotation
of start codons (D.A. Widdick, G. Chandra and T. Palmer,
unpublished). Therefore, all of the ORFs of S. scabies
were re-analysed to take account of all potential start
codons; these modified ORFs were also analysed by
TATFIND 1.4 and TatP, leading to the identification of a
further 21 potential Tat substrates, which are also listed
in Table S1. Additionally, it is known that some Tat
signals have very long n-regions that preclude their identification by TATFIND 1.4 or TatP; these programmes
have maximum preferred lengths for signal peptide
n-regions. Therefore all of the S. scabies ORFs were
truncated in silico by 30 or 60 amino acids and
re-analysed by TATFIND 1.4 and TatP resulting in the
identification of an additional five candidate Tat substrates (Table S1). Finally, all of the ORFs that were predicted to have Tat signal peptides by only one of the two
Tat signal peptide prediction programmes were sorted
manually for those that were likely to be true Tat substrates on the basis of binding a complex cofactor,
showing high homology to confirmed Tat substrates from
other organisms, or by virtue of the fact that the twin
arginine motif was highly conserved across bacterial
homologues. This added a further 14 substrates to the
manually curated list of likely Tat substrates (Table S1).
An additional four proteins were added as a result of the
outcome of the Tat-dependent reporter experiments
described below and this curated list of likely Tat substrates in S. scabies (Table S1) therefore contains a total
of 126 proteins.
Phenotype of a S. scabies DtatC deletion strain
In order to test the in silico predictions that there are many
Tat substrates encoded in the genome of S. scabies, it
was necessary to inactivate the Tat pathway. Inspection of
the genome sequence of S. scabies reveals that, like
other streptomycetes, it encodes the tatA (SCAB73591)
and tatC (SCAB73601) genes in close proximity (separated by 51 bp) and a probable tatB gene (SCAB31121) at
a distant location. Because the tatC gene encodes an
essential component of the Tat transporter (e.g. Bogsch
et al., 1998), we constructed a marked deletion of tatC as
described in Experimental Procedures.
Inactivation of the Tat pathway in S. coelicolor and
Streptomyces lividans is associated with a number of
phenotypic changes including a dispersed growth in liquid
culture (rather than the mycelial pellets that are seen for
the WT strains), failure to sporulate on solid media containing sucrose and increased fragility of the hyphae that
may reflect a cell wall defect (Schaerlaekens et al., 2004;
Widdick et al., 2006). To ascertain whether inactivation of
the Tat pathway had similar pleiotropic effects on S.
scabies, we initially assessed the growth rate of the S.
scabies WT and isogenic DtatC strains. It should be noted
that Streptomyces growth rates cannot be measured by
spectroscopic analysis as the presence of mycelial
clumps in liquid culture, coupled with the production of
cellular debris by programmed cell death, distorts results
obtained by spectroscopy (Miguélez et al., 1999). Therefore to assess growth, total cytosolic protein was
extracted from living cells and measured, as described in
Experimental Procedures.
As shown in Fig. 1A, when cultured in tryptone soya
broth (TSB) medium, the DtatC mutant strain does not
grow as well as the WT strain. While both strains reach a
peak of growth at around 28 h post inoculation (hpi), the
mutant strain showed only about half of the total protein
content per ml of culture as the WT strain. Following the
growth peak, both strains showed a subsequent decline in
the total protein content that may be related to programmed cell death (Miguélez et al., 1999). The slower
growth rate of the DtatC strain was also apparent on solid
growth media; for example, the tatC mutant strain formed
smaller colonies on yeast extract malt extract (YEME) agar
plates (Fig. 1B). It is also striking to note in Fig. 1B that the
DtatC strain lacks the brown colouration of the medium
and the hyphae that is associated with the WT strain. This
may reflect a lack of production of melanin, which is
catalysed by tyrosinase, a cofactor-containing Tat substrate (Schaerlaekens et al., 2001). Two probable tyrosinase enzymes are encoded in the S. scabies genome,
SCAB85681/85691 and SCAB59231/59241, and the
MelC1 components of both of these have candidate twin
arginine signal peptides (see Table S1). It should be noted
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 255
Fig. 1. Inactivation of the Tat pathway in S. scabies results in pleiotropic growth phenotypes.
A. 100 ml of TSB medium was inoculated with either S. scabies WT or DtatC strains, at a concentration of 100 000 spores per ml and
incubated at 30°C with shaking. At the indicated time points, 3 ¥ 1 ml samples were removed from each culture, the hyphae pelleted and total
protein content was determined using the Biorad DC Protein Assay. The error bars represent the standard error of the mean of samples taken
at indicated time points, where n = 3.
B. Spores of the indicated strains were streaked out from a fresh solid medium culture with an inoculation loop onto YEME medium and
incubated for 7 days at 30°C.
C. Spores dilutions of each strain were plated onto YEME medium and onto the same medium containing 0.01% SDS. The plates were
incubated at 30°C for 7 days.
D. Approximately 107 spores of each strain were spread on 144 cm2 DNA medium. Antibiotic discs with of 6 mm diameter were imbued with
one of the following different amounts of vancomycin in mg as indicated and plates were incubated at 30°C for 7 days.
that although the phytotoxin thaxtomin is also pigmented,
thaxtomin production was not affected by inactivating tatC
(Fig. S1). Complementation of the S. scabies DtatC strain
with an integrative plasmid harbouring the S. scabies
tatAC genes restored production of the brown pigment,
and also reversed the slow growth rates seen on solid
media, indicating that these phenotypes are directly linked
to the tatC mutation (Fig. S2).
Inactivation of the Tat pathway is often associated with
a decrease in the integrity of the bacterial cell envelope,
and in some organisms at least this is linked to an inability
to export Tat-dependent proteins involved in cell wall
remodelling (Ize et al., 2003; Caldelari et al., 2006). In E.
coli and P. syringae this has been observed as an
increased sensitivity to the presence of the detergent SDS
when the Tat pathway is inactivated. As shown in Fig. 1C,
deletion of the S. scabies tatC gene also results in an
increased sensitivity to killing by SDS, which might also
be suggestive of a Tat-linked cell envelope defect in this
organism. A possible underlying defect in the S. scabies
cell wall linked to inactivation of the Tat pathway was
investigated by determining the sensitivity of the WT and
DtatC strains to the cell wall-directed antibiotic
vancomycin. As shown in Fig. 1D, the DtatC strain was
clearly significantly more sensitive to killing by vancomycin than the WT strain, and a similar result was also seen
with a different cell wall-directed antibiotic, bacitracin
(data not shown). These observations suggest that the
cell wall of the tatC mutant strain differs from that of the
WT.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
256 M. V. Joshi et al. 䊏
Fig. 2. Close-up of protein spots that are
predominantly present in the WT extracellular
proteome in all four types of growth media.
The S. scabies WT and DtatC strains were
grown on all four of the following growth
media; IPM, OBM, R5 and SFM, and
extracellular proteins were harvested from cell
wall washes, TCA precipitated and subjected
to 2D-PAGE as described in Experimental
Procedures. Depicted are enlarged sections
of the dual channel images of the extracellular
proteomes of S. scabies WT (red colour) in
comparison to the tatC mutant strain (green
colour). The red-appearing proteins are
labelled by the SCAB number and include 28
proteins that are strongly exported in more
than one growth medium in the WT and
decreased or absent in the DtatC mutant. The
ratios for all identified proteins exported at
lower amounts by the DtatC mutant are given
in Table S2.
Analysis of the exported proteome of the S. scabies WT
and DtatC strains
We next chose to examine the Tat-dependent proteome of
S. scabies by comparison of the extracellular proteins
produced by the WT and the tatC mutant strains under
different growth conditions. As noted previously, the
complex life cycle of streptomycetes results in significant
lysis of the hyphae in liquid culture, and release of many
cytoplasmic proteins into the growth medium. Therefore,
we followed the procedure of Widdick et al. (2006), which
involves growing Streptomyces on solid media on top of
cellophane discs and washing the biomass with lithium
chloride to release cell wall-associated proteins. The
strains were cultured on four standard S. scabies growth
media – oat bran medium (OBM), instant potato mash
medium (IPM), soy-flour mannitol medium (SFM) and R5
medium (which is a more defined medium than the other
three and lacks plant-derived material), and proteins from
the cell wall washes were analysed by two-dimensional
gel electrophoresis. For each growth medium analysed,
the proteomes of the WT and tatC mutant strain were
compared using the Decodon Delta 2D software with the
WT proteins false coloured in red and the tatC proteins in
green to allow differences to be visualized. The results of
the proteome comparison for each type of growth media
are shown in Fig. S3, and close-ups of some abundant
red protein spots that are detected in the WT but are
absent or decreased in the DtatC strain in at least two
different growth media are presented in Fig. 2.
It is clear from inspecting the gels shown in Fig. S3 and
the panels in Fig. 2 that many proteins are exported at
lower amounts in the tatC mutant strain. The redappearing protein spots were identified using tryptic
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 257
digestion and MALDI-ToF mass spectrometry. In total, 103
proteins were identified – these are indicated on Figs. 2
and S3, and are listed in Table 1. The ratios of the protein
amounts for all red spots that are more highly exported in
the WT were quantified in relation to the DtatC strain and
presented in Table S2. The sequence coverages and
protein scores for all identified proteins are given in
Table S3.
Among the 103 proteins we identified, 73 proteins were
predicted to have N-terminal signal peptides. Of these, 14
were predicted to be Tat substrates by both TATFIND and
TatP (and this was increased to 19 if alternative start sites
were selected), and a further nine were predicted to be Tat
substrates by one of the two programmes. Interestingly,
these 28 predicted Tat substrates include several enzymes
that are related to phosphate starvation conditions, such
as alkaline phosphatases (SCAB77971, SCAB68191)
glycerophosphoryl diesterase (SCAB74351), and
5′-nucleotidases (SCAB68841, SCAB49491), which are
abundantly exported proteins in WT cells (Fig. 2). Furthermore, several glycosyl hydrolases (SCAB16771,
SCAB17721), cellulases (SCAB25591) and mannosidases (SCAB16551) were exported strongly in WT cells
but at significantly reduced levels in the tatC mutant strain.
In addition to these 28 predicted Tat substrates, there
were 50 signal peptide-bearing proteins that were
reduced in the extracellular proteome of the tatC mutant
but were not predicted to be Tat substrates and many of
these did not have consecutive arginines in their signal
peptides. It should be noted that the majority of exported
proteins identified as being differentially missing from the
extracellular proteome of the S. coelicolor DtatC strain
were also not Tat substrates, and it was concluded that
they were simply not produced in the DtatC strain because
of the pleiotropic effects of the mutation (Widdick et al.,
2006). It is possible that some of these differences also
relate to the different growth rates of the S. scabies WT
and DtatC strains. Finally, 30 proteins were identified in
the cell wall wash of the S. scabies WT strain that did not
appear to have N-terminal signal peptides (Table 1). It is
likely that most of these proteins are cytoplasmic and their
presence in the cell wall is due to contamination, with the
notable exception of tyrosinase (SCAB85681) which is a
known Tat substrate that lacks a signal peptide and is
transported through the Tat system by forming a complex
with a signal peptide-bearing partner protein (Chen et al.,
1992; Leu et al., 1992; Schaerlaekens et al., 2001).
Verification of S. scabies Tat-targeting signal peptides
using the agarase reporter assay
Using a combination of bioinformatic and proteomic
approaches described above we identified a large number
of candidate Tat substrates in S. scabies. A common
method for ascertaining whether a given protein is likely to
be a Tat substrate is to fuse its signal peptide to a reporter
protein and determine whether it is able to mediate export
by the Tat pathway. Recently, Widdick et al. (2006; 2008)
described the agarase reporter system as a convenient
and reliable reporter to test for Tat-dependent export.
Agarase is a Tat-exported extracellular enzyme produced
by S. coelicolor whose extracellular activity can be readily
detected by the formation of halos on agar plates after
staining with iodine. Agarase cannot be exported in an
active form by the Sec pathway, but has been shown to be
exported by the Streptomyces Tat pathway when fused to
Tat signal peptide sequences from a wide range of organisms including Gram negative bacteria, archaea and even
eukaryotes (Widdick et al., 2006; 2008). Because the size
of the halo is related to the amount of agarase secreted,
the assay is also semi-quantitative.
We therefore selected 35 proteins that had been identified from the extracellular proteome of the WT strain,
including 13 from Table 1 that were predicted to be Tat
substrates by both TATFIND 1.4 and TatP, three identified
by TATFIND 1.4 only, two identified by TatP only, a further
16 that had recognizable signal peptides that were not
predicted to be Tat-targeting by either prediction programme, and one that as presented had no apparent
signal peptide but if an upstream start site was used it had
a good candidate Tat signal peptide. The signal peptides
from each of these proteins were fused in-frame to the
mature region of agarase. The constructs are designed
such that each clone carries the dagA ribosome binding
site, with identical spacing between the ribosome-binding
site and the start codon (Widdick et al., 2006; 2008). The
recombinant proteins were expressed in both S. lividans
WT and DtatC strains, and scored for agarase activity. In
total, 14 of the 35 signal peptides were able to mediate
Tat-dependent export of agarase (Fig. 3A) and these proteins are listed in Table 2.
Of the proteins that had signal peptides that could
export agarase, ten of them were predicted to have Tat
signal peptides by both programmes, and a further two by
one or other programme. The remaining two (SCAB15581
and SCAB68191) did not have Tat signal peptides on the
ORFs as called, but if plausible upstream start codons
were used both had very good signal peptides that were
recognized as Tat-targeting by both TATFIND 1.4 and
TatP. This observation strongly suggests that the start
codons of these two proteins have been mis-annotated
(http://www.sanger.ac.uk/Projects/S_scabies/).
Interestingly, for two of the signal peptides that were
able to mediate export of agarase, no agarase activity
was detected if alternative versions of these signal peptides were tested. For SCAB15581, if the n-region of the
signal peptide was truncated by selecting an alternative
start codon closer to the twin arginine motif, no agarase
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
258 M. V. Joshi et al. 䊏
Table 1. Proteins that are predominantly found in the extracellular fraction of the S. scabies wild-type strain following 2D gel analysis.
Protein
Growth medium detected
Pass TATFIND 1.4 and TatP
SCAB08951
R5
SCAB15571
IPM, OBM, R5
SCAB16551
OBM, SFM
SCAB16651
IPM
SCAB25591
IPM, OBM, R5, SFM
SCAB38731
OBM
SCAB47131
IPM, OBM
SCAB59671
R5
SCAB63891
IPM, SFM
SCAB68841
IPM, OBM, R5, SFM
SCAB74351
IPM, OBM, R5, SFM
SCAB75721
IPM, OBM, R5, SFM
SCAB77971
IPM, OBM, R5, SFM
SCAB81841
OBM
Pass TATFIND 1.4 only
SCAB18501
IPM, OBM, R5, SFM
SCAB27931
SFM
SCAB37611
IPM, OBM, R5, SFM
OBM, R5
SCAB57891c
SCAB82451
SFM
Pass TatP only
SCAB08221
IPM, OBM, R5, SFM
SCAB64081
IPM, SFM
SCAB78761
OBM
SCAB84861
R5
Pass Signal P but not TATFIND 1.4 or TatP
SCAB01191
IPM
SCAB04761
R5
SCAB05351
R5
SCAB07651
IPM
SCAB08871
IPM, OBM
SCAB13321
R5
SCAB13551
SFM
SCAB14911
SFM
SCAB16431
OBM
SCAB16711
IPM, OBM, R5, SFM
SCAB16721
IPM, OBM, R5, SFM
SCAB17001
IPM, SFM
SCAB17571
IPM
SCAB18081
IPM, SFM
SCAB18661
SFM
SCAB19481
IPM
SCAB19491
SFM
SCAB19941
SFM
SCAB24731
OBM
SCAB24891
SFM
SCAB26841
SFM
SCAB27771
OBM
SCAB31531
IPM, R5
SCAB33981
SCAB36371
SCAB41181
SCAB43901
SCAB44001
SCAB44161
SCAB44541
SCAB44691
SCAB45141
SCAB47841
SCAB49491
SCAB56441
SCAB58251
SCAB59651
SCAB66031
SCAB68191c
SFM
IPM
SFM
IPM
SFM
R5, SFM
IPM, OBM
IPM, OBM, SFM
SFM
R5, SFM
IPM, OBM, R5, SFM
IPM
SFM
SFM
SFM
IPM, OBM, R5, SFM
Putative function
Agarase testa
ABC-type Fe3+ transport system, periplasmic component
Putative secreted phosphoesterase
Putative mannosidase
Putative secreted protein
Putative secreted cellulase
Putative secreted beta-lactamase
Peptidase family M23/M37 protein
Putative hydrolase
Putative secreted transport-associated protein
Putative 5′ nucleotidase
Glycerophosphoryl diester phosphodiesterase
Putative secreted protein
Putative secreted phosphatase (fragment)
Putative secreted hydrolase
Pass
Pass
ND
Pass
Pass
Pass
Failb
Pass
Failb
Pass
Pass
Pass
Failb
Pass
Alpha-N-acetylglucosaminidase
Putative neutral zinc metalloprotease
Putative secreted aminopeptidase
Conserved hypothetical protein
Putative secreted protein
Pass
Failb
Fail
ND
ND
Conserved hypothetical protein
Putative secreted protein
Putative secreted protein
Putative secreted amidase
Pass
Fail
ND
ND
Putative secreted protein
Putative secreted protein
Substrate-binding component of ABC transporter
Neutral zinc metalloprotease
Secreted endoglucanase
S15 non-peptidase homologue family
Conserved hypothetical protein
Putative secreted protein
Putative possible cellulase CELA1
Putative secreted glycosyl hydrolase
Putative secreted glycosyl hydrolase
Secreted cellulase
Putative secreted peptidase
Gamma-glutamyltranspeptidase
Putative serine protease
RTX-family exoprotein
Putative probable exported protein
Cyclase
Putative secreted aminopepetidase
Glutamate-binding protein
Putative serine protease
Putative secreted protein
Putative BldKB-like transport system extracellular
solute-binding protein
Hypothetical protein 2SCK36.08
Putative xylanase/cellulase
Conserved hypothetical protein
Putative secreted hydrolase
Putative secreted protein
Hydrolase
Putative secreted protein
Putative secreted protein
D-alanyl-D-alanine carboxypeptidase
Lipoprotein
Putative secreted 5′-nucleotidase
Putative secreted protease
Putative secreted tripeptidyl aminopeptidase
Putative secreted serine protease
Putative glycosyl hydrolase
Putative secreted alkaline phosphatase
ND
ND
Fail
Fail
ND
ND
Fail
ND
ND
Failb
Fail
ND
Fail
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Fail
ND
Fail
ND
ND
ND
Fail
ND
ND
ND
ND
ND
Fail
ND
Passb
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 259
Table 1. cont.
Protein
Growth medium detected
Putative function
Agarase testa
SCAB68931
SCAB69691
SCAB72441
SCAB72781
SCAB74081
SCAB78431
SCAB80071
SCAB84971c
SCAB85291
SCAB90091
SCAB90101
No signal peptide
SCAB15581c
SCAB17551
SCAB20121
SCAB21191
SCAB25251
SCAB26011
SCAB31831
SCAB36931
SCAB37201
SCAB39491
SCAB39811
SCAB41891
SCAB44821c
SCAB45751
SCAB50441
SCAB51341
SCAB51541
SCAB54151
SCAB55881
SCAB57721
SCAB58791
SCAB59311
SCAB59871
SCAB62141
SCAB64141
SCAB64311
SCAB67061
SCAB69391
SCAB71391
SCAB85681d
SFM
IPM
IPM, SFM
IPM
IPM, OBM, R5
SFM
SFM
IPM
IPM
IPM, OBM, SFM
SFM
Branched chain amino acid-binding protein
Zinc-binding carboxypeptidase
Putative membrane protein
Penicillin acylase
Putative secreted protein
Secreted tripeptidylaminopeptidase
Putative secreted protein
Conserved hypothetical protein
Conserved hypothetical protein
Secreted cellulase
Secreted cellulase
ND
Fail
Failb
ND
ND
Fail
ND
Fail
ND
fail
ND
IPM, OBM, R5, SFM
IPM
R5, SFM
IPM, SFM
OBM
OBM
R5
R5
OBM
OBM, R5, SFM
R5
R5
IPM, OBM, R5, SFM
R5
R5
OBM, R5
SFM
OBM
IPM
OBM
R5
R5
SFM
SFM
SFM
SFM
SFM
R5
OBM
R5
Conserved hypothetical protein
Conserved hypothetical protein
Putative germacradienol synthase
Hypothetical protein
Guanosine pentaphosphate synthetase
Elongation factor Ts
Cytochrome P-450 hydroxylase
50S ribosomal protein L4
50S ribosomal protein L7/L12
Clp-family ATP-binding protease
Conserved hypothetical protein
Adenylosuccinate synthetase
Conserved hypothetical protein
DNA gyrase subunit A
Chaperonin 2
Citrate synthase
Calcium binding protein
Ribosomal L25p family protein
Transcriptional regulator
Cellobiose hydrolase
Citrate synthase
ABC transporter ATP-binding protein
GntR family DNA-binding regulator
Pyruvate phosphate dikinase
Acyl carrier protein
Conserved hypothetical protein
Dihydrolipoamide dehydrogenase
Conserved hypothetical protein
Conserved hypothetical protein
Putative tyrosinase MelC2
Passb
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
a. Signal peptide sequences (listed in Table S5) of the indicated proteins were fused to the mature region of agarase and their ability to export
agarase in a Tat-dependent manner was determined as described in the text.
b. Two variants of each of these signal peptides were tested (see Table S5).
c. The N-termini of these proteins are recognized as Tat signal peptides by both TATFIND 1.4 and TatP if alternative start sites are selected.
d. This protein lacks an N-terminal signal peptide but is a Tat substrate by virtue of interacting with a Tat signal peptide-bearing partner protein,
MelC1 (Leu et al., 1992; Schaerlaekens et al., 2001).
ND, not determined.
activity was detected. For SCAB68191, which does not
have a twin arginine motif in the ORF as called, two
alternative start codons were selected that both produced
peptides containing a twin arginine motif but with different
lengths of n-region. Again only the peptide with the longer
n-region gave any detectable agarase activity. This indicates that the choice of start codon may be critical to
whether a given signal peptide is able to mediate agarase
export.
Surprisingly, three of the signal peptides in Table 2 that
were predicted to be Tat-targeting by both TATFIND 1.4
and TatP did not mediate detectable export of agarase.
The reason for this is not clear; however, given the observation that choice of start site can significantly affect the
extracellular agarase activity, it is possible that incorrect
start sites were selected for these signal peptides. This
may be particularly relevant for SCAB77971, which is a
predicted PhoD family phosphatase that are known to
have very long and variable signal peptide n-regions
(Widdick et al., 2006; 2008). Alternatively, it is possible
that these are either not Tat-targeting signals or that they
are genuine Tat-targeting signals, but that they do not
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
260 M. V. Joshi et al. 䊏
Fig. 3. Export of agarase mediated by S.
scabies signal peptides. The x-axis shows a
range of signal peptides from S. scabies from
either (A) cell wall-associated proteins that
were differentially absent from the DtatC strain
(selected from the list in Table 2) or (B) that
were predicted to be Tat signal peptides by
bioinformatic analysis (listed in Table S1). The
y-axis gives a measure of agarase export
from each fusion protein, compared with
agarase bearing its native signal peptide
(DagA) cloned in the same manner (set at
100%). Each construct carries the native
agarase promoter and ribosome binding site.
The error bars represent the standard error of
the mean, where n = 3.
A. Signal peptides from the following proteins
listed in Table 1 were also tested and found to
be negative in this assay: SCAB07651,
SCAB13551, SCAB16711 (two variants),
SCAB16721, SCAB17571, SCAB27931 (two
variants), SCAB36371, SCAB37611,
SCAB43901, SCAB44691, SCAB47131 (two
variants), SCAB59651, SCAB63891 (two
variants), SCAB64081, SCAB69691,
SCAB72441 (two variants), SCAB77971 (two
variants), SCAB78431, SCAB84971,
SCAB90091.
B. Two variants of the SCAB63071 signal
peptide were able to mediate export of
agarase to differing degrees – both are shown
in the Figure. Signal peptides from the
following proteins listed in Table S1 were also
tested and found to be negative in this assay:
SCAB08301, SCAB31461, SCAB34181,
SCAB70581 and SCAB78851. The exact
amino acid sequences of each of the
sequences tested in these assays can be
found in Table S5.
direct the Tat-dependent export of agarase for some
unknown reason.
In addition to testing signal peptides from proteins
shown in the proteomic study to be predominantly present
in the extracellular protein fraction of the WT strain, we
also tested a further 38 signal peptides that were predicted to be Tat-targeting based on the bioinformatic
analysis of the S. scabies genome sequence described
above (and listed in Table S1). These 38 proteins are
made up of 28 proteins that were predicted to have Tat
signal peptides based on the fact that they were recog-
nized by TATFIND 1.4 and TatP, six that were recognized
by TATFIND 1.4 and TatP following reassignment of the
start codon, two that were recognized by both prediction
programmes after in silico truncation of the N-terminal part
of the ORF, one protein that was predicted to have a Tat
signal peptide by TATFIND 1.4 only, and one that was
predicted by TatP only.
As shown in Fig. 3B, signal peptides derived from 33 of
the proteins tested were able to mediate export of
agarase and we therefore conclude that these proteins
are Tat substrates. The proteins that bear these signals
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 261
Table 2. Proteins from S. scabies with signal peptides that are able to mediate the Tat-dependent export of agarase.
Signal peptide tested
Putative function
Protein detected by proteomics
SCAB08221
SCAB08951
SCAB15571
SCAB15581
SCAB16651
SCAB18501
SCAB25591
SCAB38731
SCAB59671
SCAB68191
SCAB68841
SCAB74351
SCAB75721
SCAB81841
Proteins predicted by bioinformatics
SCAB00601
SCAB03871
SCAB04961
SCAB06471
SCAB09381
SCAB09591
SCAB10131
SCAB13731
SCAB15711
SCAB16951
SCAB19551
SCAB25621
SCAB37061
SCAB46711
SCAB48941
SCAB57331
SCAB57661
SCAB63071
SCAB66251
SCAB73161
SCAB74641
SCAB74681
SCAB77391
SCAB77401
SCAB77511
SCAB78741
SCAB79011
SCAB79571
SCAB80581
SCAB81041
SCAB81901
SCAB88501
SCAB89661
Putative phosphoesterase
ABC-type Fe3+ transport system, periplasmic component
Putative secreted phosphoesterase
Possible histidinol phosphatase (PHP family)
Putative L-xylulose-5-phosphate 3-epimerase
Alpha-N-acetyl glucosaminidase
Endo-1,4-beta-glucanase
Beta-lactamase
Cyclic 3′,5′-adenosine monophosphate phosphodiesterase
Phospholipase D precursor
2′,3′-cyclic-nucleotide 2′-phosphodiesterase
Glycerophosphoryl diester phosphodiesterase
A subfamily of peptidase family C39
Alpha-L-rhamnosidase
Hypothetical protein
Glycosyl hydrolase domain followed by ricin domain
Endo-xylanase
Putative alpha-L-fucosidase
Hypothetical protein
Alpha-L-fucosidase
Glycosyl hydrolase domain followed by ricin domain
Phospholipase C
Putative endo-1,3-beta-glucanase
Hypothetical protein
Rhamnogalacturonase B precursor
Lipoprotein
Putative secreted protein
Protocatechuate dioxygenase
ATP-dependent nuclease subunit B-like
No homologues in database
Putative multiple sugar ABC transporter solute-binding protein
Putative gluconolactonase
Iron-dependent peroxidase
Gluconolactonase
Lipoprotein
Hypothetical protein
Hypothetical Protein
Glycosyl hydrolase
Beta-galactosidase
Putative secreted protein
Lipase/acylhydrolase, putative
Large secreted protein
Amine oxidase
Putative ABC transporter periplasmic binding protein
Peptide ABC transporter peptide-binding protein
Rhamnogalacturonase B precursor
Putative factor C protein precursor
The amino sequences of each signal peptide tested are given in Table S5.
are also listed in Table 2. Of the signal peptides tested in
this section, two variants of the SCAB00601 signal
peptide were tested, of which only the clone that encoded
a signal peptide with the longer n-region (pSM29b, see
Table S5), gave any detectable activity. Likewise, three
variants of the SCAB63071 signal peptide were tested. Of
these, the agarase-producing clone that encoded the
longest n-region (pSM35a) failed to give detectable
agarase activity while the remaining two both gave detectable agarase export but at different levels. This indicates
that the features of the cloned DNA sequence and/or the
sequence of the signal peptide tested in this assay may
have a profound effect on the outcome of the reporter
assay.
Five of the signal peptides identified as Tat-targeting
by bioinformatic means (SCAB08301, SCAB31461,
SCAB34181, SCAB70581 and SCAB78851) did not
promote the export of agarase. Again the reasons for this
is not clear – it may be related to the fact that inappropriate start codons were selected for these signal peptides,
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
262 M. V. Joshi et al. 䊏
Fig. 4. Reduced disease severity of
Arabidopsis seedlings inoculated with the
DtatC mutant strain. Arabidopsis seedlings
(7 days old) were inoculated with spores
(corresponding to 1 ¥ 106 cfu) of the S.
scabies WT (87.22) and DtatC mutant (DtatC)
strains, keeping un-inoculated seedlings as
controls. The plants were grown for a further
21–28 days, and the images are taken at 28
days post inoculation.
or that they were genuine Tat signal peptides but were
unable to mediate detectable export of agarase. Alternatively, it is possible that despite the fact they were predicted to be Tat signal peptides by both TATFIND and TatP
they are actually Sec-targeting. In this context it is interesting to note that SCAB31461 is the S. scabies homologue of S. coelicolor BldKB that has a KR rather than an
RR motif in its signal peptide and is also unable to
mediate export of agarase (Widdick et al., 2006). This
would be consistent with the idea that this protein may be
a Sec substrate in Streptomyces.
In total, the agarase reporter assay has identified 47
Tat-targeting sequences, four of which were predicted to
be Tat-targeting by only one of the two prediction programmes, and we conclude that it is highly likely that
these proteins are Tat substrates in S. scabies.
The Tat secretion system contributes to the virulence of
S. scabies
The demonstrated role of Tat-secreted proteins in the
virulence of other microbial pathogens (reviewed in De
Buck et al., 2008) and the existence of proteins in the S.
scabies genome that have homologues in pathogenic
microbes, suggested a likely role for Tat secretion in
virulence. We therefore investigated the virulence of the
DtatC mutant, relative to the WT strain, on the model plant
Arabidopsis. S. scabies is a broad host range pathogen
that causes root rot on both model plant species and
agricultural crops (reviewed in Loria et al., 2006). The root
tips of Arabidopsis (Col-O) seedlings were inoculated with
S. scabies spores from either the WT (WT; 87-22) or the
DtatC mutant, and seedling growth was monitored weekly.
As shown in Fig. 4, infection of seedlings by the WT
strain caused root stunting and necrosis; secondary roots
were killed soon after emergence from the taproot. In
addition, the leaves and shoots were stunted, chlorotic
and necrotic; all of the plants inoculated with the WT strain
were dead within 21–30 days post inoculation. By contrast, plants inoculated with the DtatC mutant strain grew
vigorously and were similar to the un-inoculated control in
size and colour (Fig. 4). Interestingly, the DtatC mutant
grew extensively on the roots as yellow substrate mycelium, without noticeable sporulation, but did not appear to
necrotize the colonized tissue. Plants inoculated with the
DtatC mutant did, however, differ from the un-inoculated
control in root morphology, particularly root-branching
pattern (Fig. 4).
Given the dramatic virulence phenotype of the DtatC
mutant, we carried out a time course study of Arabidopsis
root colonization using confocal scanning microscopy and
enhanced green fluorescent protein (EGFP)-labelled S.
scabies WT and DtatC strains. To facilitate observation of
microbial growth, plants were grown hydroponically. The
WT strain attached to the root tip and began to colonize
plant tissues within 24 h after spores were added to the
hydroponic medium (Fig. 5A). Within 48 hpi the WT strain
was aggressively colonizing the root at the zone of cell
differentiation (Fig. 5B) and at lateral meristems (Fig. 5C).
Intercellular colonization was evident at the root tip and
at the point of secondary root emergence by 72 hpi
(Fig. 5D). By contrast, colonization of root tissue by the
DtatC mutant was greatly delayed and limited in scope.
The mutant was not able to attach to the root tip during the
first 48 hpi; however, there were loosely attached colonies
at the root elongation zone (Fig. 5E), and minimal colonization and restricted growth at the differentiation and
elongation zones (Fig. 5F–G), and at lateral meristems
(Fig. 5H) after 72–96 hpi. Intercellular colonization was
delayed until 96–120 hpi (Fig. 5I–J) and was limited in
scope.
Individual Tat-secreted proteins contribute to the
virulence of S. scabies
Because the DtatC strain was essentially avirulent on
Arabidopsis seedlings, we sought to address the contribution of individual candidate Tat substrates to host–
pathogen interactions. Fourteen individual strains were
constructed that were deleted for genes encoding the
putative or confirmed Tat substrate proteins listed in
Table 3. The strains were then assessed for virulence
using Arabidopsis seedlings as a host. To this end, Arabidopsis seedlings were grown on Murashige and Skoog
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 263
Fig. 5. Micrographs of Arabidopsis roots colonized by EGFP-labelled S. scabies strains. Colonization and disease progression was
monitored on hydroponically grown Arabidopsis roots after inoculation with EGFP labelled S. scabies strains. (A–D) Inoculation with the WT
strain 87-22, and (E–J) inoculation with the DtatC mutant strain. Colonies of the WT strain are shown (A) 24 h post inoculation (hpi) showing
cells at and near the root tip (red arrow); (B and C) 48 hpi showing colonization of (B) the root at the zone of cell differentiation and (C) at a
lateral meristem; and (D) 72 hpi at a lateral root meristem – intercellular growth is indicated by the red arrows. Colonies of the tatC mutant
strain are shown (E) at the root tip zone during the first 24–48 hpi; the arrow indicates loosely attached mycelium at the elongation zone;
(F–H) after 72–96 hpi; (F) at the differentiation zone (G) at elongation zones and (H) at a lateral meristem; and (I and J) at 96–120 hpi. (K and
L) Un-inoculated control seedlings showing (K) normal root growth, and (L) loose cells surrounding the root cap region. Images were taken
using a Leica TCS SP5 confocal microscope. Size bars represent 50 mm.
(MS) medium and inoculated with spore suspensions of
the WT or with mutant strains containing deletions in
genes encoding candidate Tat substrates using methods
just described. Of the 14 individual mutant strains tested,
seven were reduced in virulence relative to the WT strain
(Fig. 6).
Gene deletions in SCAB03871 and SCAB10131 were
among those coding for Tat substrates that have a
reduced virulence phenotype in the Arabidopsis seedling
assay. Interestingly, these genes encode paralogues with
predicted glycosyl hydrolase and ricin-like domains. In
spite of the similarity in their amino acid sequences, each
of these proteins appears to make a large and unique
contribution to virulence, based on the phenotypes of
the individual mutants (Fig. 6). Furthermore, proteins
encoded by SCAB03871 and SCAB10131 have homologues in fungal plant pathogens such as Chaetomium
globosum (CHGG_02005) and Phaeosphaeria nodorum
(SNOG_01152) respectively.
SCAB06471 encodes a putative alpha-L-fucosidase;
the cognate deletion mutant was compromised in virulence (Fig. 6). Alpha-L-fucosidases are responsible for
processing of fucosylated glycoconjugates that play a role
in a wide variety of biological processes. Interestingly,
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
264 M. V. Joshi et al. 䊏
Table 3. Potential Tat-dependent virulence factors encoded in the genome of S. scabies.
Gene ID
Putative function
SCAB03871
Glycosyl hydrolase domain
followed by ricin domain
SCAB06471
Putative alpha-L-fucosidase
SCAB08951
ABC-type Fe3+ transport
system, periplasmic
component
Glycosyl hydrolase domain
followed by ricin domain
SCAB10131
SCAB18791
Hypothetical secreted
protein
SCAB58511
Hypothetical protein.
SCAB58531
Hypothetical protein
SCAB70581
Hydrolase of the alpha/beta
superfamily
SCAB76661
Rare lipoprotein A
SCAB77391
Conserved hypothetical
protein containing a
galactose-binding domain
Putative glycosyl hydrolase
SCAB77401
SCAB80581
Putative FAD-containing
amine oxidase
SCAB81041
Putative
spermidine/putrescine
transporter
peptide-binding protein
Alpha-L-rhamnosidase
SCAB81841
Homologs with highest similarity
(identity/similarity)
Locus_tag
Streptomyces sviceus (89/93)
Micromonospora aurantiaca (76/85)
Actinosynnema mirum (71/81)
Streptomyces sviceus (82/90)
Streptomyces viridochromogenes(83/89)
Streptomyces hygroscopicus (57/70)
Acidovorax delafieldii (60/74)
Phytophthora infestans (51/67)
Pectobacterium carotovorum (48/63)
Streptomyces sviceus (84/92)
Micromonospora sp. L5 (75/83)
Actinosynnema mirum (70/80)
Bacteroides cellulosilyticus (39/57)
Pectobacterium carotovorum (34/50)
Ralstonia solanacearum (34/49)
Pseudomonas putida (27/45)
Nitrobacter winogradskyi (27/43)
Clostridium botulinum A3 (25/42)
Nitrobacter winogradskyi (29/45)
Clostridium botulinum Ba4 (26/43)
Pseudomonas putida (29/45)
Streptomyces sviceus (76/82)
Streptomyces griseoflavus (76/83)
Streptomyces ghanaensis (76/83)
Streptomyces sviceus (54/65)
Micromonospora sp. (66/79)
Streptomyces viridochromogenes (72/82)
Bacteroides cellulosilyticus (36/53)
Pectobacterium carotovorum (30/47)
Ralstonia solanacearum (29/44)
Bacteroides ovatus (45/63)
Dickeya dadantii (36/52)
Ralstonia solanacearum (36/52)
Streptomyces pristinaespiralis (78/87)
Streptomyces sp. (76/85)
Nocardia farcinica (54/65)
Streptomyces sviceus (58/72)
SSEG_02060
MicauDRAFT_3913
Amir_3107
SSEG_02413
SvirD4_010100003282
ShygA5_010100010179
AcdelDRAFT_2816
PITG_03258
PC1_1331
SSEG_02060
ML5DRAFT_2741
Amir_3107
BACCELL_04049
PC1_0414
RSIPO_04319
PP_2006
Nwi_0886
CLK_A0070
Nwi_0886
CLJ_0010
PP_2006
SSEG_08785
SgriT_010100028100
SghaA1_010100027639
SSEG_04607
MCAG_02629
SvirD4_010100009886
BACCELL_04049
PC1_0414
RRSL_03557
BACOVA_01686
Dd586_1768
RSMK02989
SSDG_02025
StreC_010100030349
nfa3040
SSEG_09980
Streptomyces hygroscopicus(56/72)
ShygA5_010100044540
Sphaerobacter thermophilus (36/53)
Sthe_3297
Clostridium leptum (35/54)
Victivallis vadensis (37/51)
Lactobacillus rhamnosus (33/48)
CLOLEP_02977
Vvad_PD1953
LRH_00532
Agarase
testa
Virulence
phenotypeb
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
ND
No
ND
No
ND
No
ND
No
ND
No
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
a. Signal peptide sequences (listed in Table S5) of the indicated proteins were fused to the mature region of agarase and their ability to export
agarase in a Tat-dependent manner was determined as described in the text.
b. Virulence phenotype of knock-out strains tested using the Arabidposis thaliana infection model.
ND, not determined.
SCAB06471 has homologues in the fungal plant pathogens such as Gibberella zeae (FG11254.1) and Magnaporthe grisea (MGG_03257). Furthermore, all three of
these proteins are putative glycosidases and the cognate
mutants have similar virulence phenotypes; in all cases
plant growth was substantially greater than those inoculated with the WT strain, root necrosis was lacking, and
the mutant strains grew luxuriously on Arabidopsis roots
(Fig. 6). It is tempting to speculate that these proteins
have a role in penetration of the plant cell wall.
The SCAB77391 mutant strain was slightly reduced in
virulence, relative to WT (Fig. 6). The encoded protein is
predicted to contain a galactose-binding domain. It is possible that this domain recognizes specific carbohydrate
moieties on the host cell surface; in that way the protein
might function as a lectin, which could enhance host
binding or recognition. The limited virulence phenotype,
however, suggests a function that is either redundant in
the genome or has only a minor role in host–pathogen
interactions.
The SCAB81841 mutant strain has a dramatic avirulence phenotype; plants inoculated with this mutant were
comparable to the non-inoculated control, except for leaf
chlorosis and a delay in flowering. The predicted function
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 265
Fig. 6. Tat substrates are essential for the
virulence of S. scabies. Deletion mutants of
putative Tat-secreted virulence proteins are
constructed using the ReDirect gene
disruption protocol and compared to WT
87-22 for disease severity. Mutants
SCAB77391, SCAB80581, SCAB81041,
SCAB81841, SCAB03871, SCAB06471 and
SCAB10131 show suppression of disease
severity in terms of root necrosis and aerial
growth.
for the encoded protein is an alpha-L-rhamnosidase,
which hydrolyses the terminal non-reducing alpha-Lrhamnose residues in alpha-L-rhamnosides. Given the
severe defect in virulence, it seems unlikely that this
enzyme’s primary function is to hydrolyse plant biomass.
Some rhamnosides are biologically active, including antimicrobial saponins (Morrissey et al., 2000). Interestingly,
the SCAB81841 mutant grew poorly on Arabidopsis roots
(Fig. 6), consistent with a role in the degradation of an
antimicrobial molecule.
The SCAB81041 deletion mutant showed a moderate
virulence phenotype. The cognate protein encodes a
putative spermidine/putrescine transporter peptidebinding protein (Table 3).
SCAB80581 encodes a putative FAD-containing amine
oxidase. Proteins in this family additionally contain a
second domain that is responsible for specifically binding
a substrate and catalysing a particular enzymatic
reaction. The deletion mutant has a moderate phenotype
as shown in Fig. 6.
Discussion
In this study we have characterized the Tat secretome of
the plant pathogen, S. scabies. In keeping with analyses
of other streptomycete genomes, in silico analysis using
Tat substrate prediction programmes TATFIND 1.4 and
TatP predicts many candidate Tat substrates are encoded
by S. scabies. Using the agarase reporter assay, we show
here that the signal peptides of 47 candidate S. scabies
Tat substrates were able to mediate Tat-dependent
export, strongly suggesting that these represent bona fide
Tat substrates. Interestingly, two of the Tat signal peptides
identified in this study were not recognized by TATFIND
1.4 because they have hydrophobic amino acids in the +1
position of the twin arginine motif. Stanley et al. (2000)
previously noted that the amino acid residue at this position is usually polar, but the work presented here shows
that valine and leucine can both be tolerated at this
position. Furthermore, additional highly likely Tatdependent proteins listed in Tables S1 and S2 also have
signal peptides that are not recognized by TATFIND 1.4
due to the presence of His, Ile or Phe at the +1 position.
Further work is required to ascertain whether these additional ‘allowed’ residues at the +1 position are specific to
Streptomyces proteins, or whether they reflect a general
tolerance for the nature of the amino acid at this position
of Tat signal peptides. None-the-less, taken together our
analysis supports the contention that there are well in
excess of 100 Tat substrates in S. scabies.
Consistent with the notion that the S. scabies Tat
pathway is a major route of protein secretion, inactivation
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
266 M. V. Joshi et al. 䊏
of the Tat pathway was associated with a number of
phenotypes. In particular, growth rate of the DtatC strain
was reduced relative to WT and the mutant displayed a cell
wall defect, a phenotype that is also linked to Tat pathway
inactivation in other bacteria (e.g. Ize et al., 2003; Caldelari
et al., 2006; Widdick et al., 2006). The DtatC strain was
also slow to sporulate on some types of solid media.
Interestingly, the signal peptide of the S. scabies homologue of the extracellular signalling protein factor C
(SCAB89661), that is involved in cellular differentiation in
some species of Streptomyces (Birkó et al., 1999) was
shown to mediate Tat-dependent export of agarase
(Fig. 3B). This indicates that the S. scabies factor C homologue is a Tat substrate. The twin arginine signal peptides
of streptomycete factor C proteins are highly conserved
and the mature portion of the protein shows strong
sequence identity to secreted proteins encoded by mycelial fungi, raising the possibility that this DNA coding this
protein was acquired by lateral gene transfer (Chater et al.,
2010).
Analysis of the extracellular proteomes of the WT and
DtatC strains showed that in total 73 predicted secreted
proteins were present at decreased amounts or absent in
the extracellular proteome of the DtatC mutant. Using the
agarase reporter assay, 14 predicted Tat substrates
detected by proteomics as being abundantly exported in
more than one growth condition in WT cells and strongly
reduced in the DtatC mutant were verified to have Tattargeting signal peptides. However, the majority of proteins that were apparently absent from the DtatC cell wall
wash fraction are unlikely to be Tat substrates, again
underscoring the pleiotropic nature of the mutation.
Bacteria in symbiotic associations with eukaryotes rely
on secreted molecules to manipulate host cell physiology.
In the broadest definition, the small molecules and proteins that alter host cell structure and function are referred
to as ‘effectors’ (Hogenhout et al., 2009). In S. scabies
thaxtomin and a coronatine-like molecule meet the criteria
of small molecule effectors. Thaxtomin acts as a cellulose
synthesis inhibitor, through an undefined mechanism, and
is believed to aid in penetration of plant tissue (Scheible
et al., 2003; Bischoff et al., 2009). A coronatine-like molecule likely serves as a molecular mimic of the plant
defence signalling molecule, jasmonic acid (Bignell et al.,
2010). The secreted proteins Nec1 and TomA are authentic and putative effectors, respectively, in S. scabies but
other effector proteins had not previously been identified.
Effector proteins interact with host cells in many ways
and the host cell targets of most effectors are not known
(Cunnac et al., 2009); however, molecular mimicry and
modification of host cell molecules are two broad categories of virulence mechanisms. Activity, abundance and
localization of eukaryotic proteins, including those
involved in host defence, can be regulated through attach-
ment of ubiquitin or ubiquitin-like molecules. Many pathogens produce proteins that mimic components of the host
ubiquitination machinery (reviewed in Spallek et al.,
2009). Other defence and development signalling cascades are targeted by effector proteins, such as Rho
GTPase regulators, cytoskeletal modulators and host
innate immunity (Shames et al., 2009). Under the broad
definition of protein effectors suggested by Hogenhout
et al. (2009), secreted enzymes involved in degradation of
host cell molecules, particularly plant cell wall components, are also considered to be effector proteins.
Microbial pathogens can undergo rapid evolution
through the acquisition of pathogenicity islands (PAIs),
which are regions in the genome encoding multiple
virulence-associated genes (Lovell et al., 2009). In Gram
negative pathogens, PAIs often have a modular configuration and include genes encoding both the type III secretion apparatus and effector proteins that use that pathway
(Lindeberg et al., 2009). The type III secretion apparatus
provides a very effective mechanism for introduction of
effector proteins directly into host cells, but is lacking in
Gram positive bacteria. Gram positive pathogens, such as
S. scabies, have PAIs (Kers et al., 2005) but must rely on
general protein secretion systems, such as Tat and Sec,
which deliver effector proteins outside of the bacterial cell;
genes encoding effector proteins are not typically clustered with those encoding protein secretion machineries
in these pathogens.
Translocation of putative virulence factors is Tatdependent in many microbial pathogens including those
that contain type III secretion systems (reviewed in De
Buck et al., 2008). Given the importance of the Tat system
in the genus Streptomyces, it was not surprising to find that
inactivation of the Tat secretion machinery in S. scabies
resulted in an essentially avirulent phenotype (Figs 4 and
5). The tatC mutant strain was slow to colonize and invade
rapidly expanding root tissue, relative to the WT strain,
which likely was the result of several deficiencies. However
the avirulence phenotype of the S. scabies tatC mutant
strain was not due to a deficiency in thaxtomin production
(Fig. S1). Regardless, the dramatic virulence phenotype of
the DtatC mutant did suggest that a subset of Tat substrates
has virulence functions. Furthermore, the combination of
bioinformatic and proteomic analysis identified a large
number of putative Tat substrates in S. scabies, of which
more than one-third are associated with stress responses
or virulence in other bacterial or fungal pathogens. This
group of putative Tat-secreted virulence proteins includes
putative lipoproteins, ABC transporters, phospholipases/
phosphoesterases, beta-lactamase and proteins involved
in Fe homeostasis (see Tables 3 and S1).
To confirm that Tat substrates are involved in infection,
we inactivated genes encoding 14 candidate Tat substrates, from which strains inactivated for production of
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 267
seven confirmed Tat substrates showed a reduction in
virulence (Fig. 6). Because the growth rate of these
mutant strains was comparable to the WT (shown in
Fig. S4 for the SCAB03871, SCAB06471, SCAB10131
and SCAB77391 knockout strains), it can be concluded
that the Tat pathway secretes multiple virulence factors in
S. scabies.
Inspection of the list of these authentic virulence proteins
using bioinformatics reveals some interesting ecological
and phylogenetic associations. Out of the seven Tatsecreted proteins affecting virulence, SCAB77391 and
SCAB81841 are conserved in S. turgidiscabies, but the
encoding genes are absent from the genomes of nonpathogenic streptomycetes for which genome sequence
data are available at this time (data not presented).
SCAB03871, SCAB06471 and SCAB10131 all contain
homologues in fungal plant pathogens and are all predicted to interact with glycans (Table 3). Surprisingly, the
amino acid sequences of the Tat secretion signals of these
three proteins are highly conserved, possibly suggesting a
common origin. SCAB03871 and its paralogue (confirmed
by reciprocal BLAST analysis) SCAB10131 have glycosyl
hydrolase and ricin-like domains. The closest homologues
to these proteins are a few closely related saprophytic
actinomycetes, where the twin arginine signal peptide is
always conserved. Outside of the actinomycetes, the
closest homologues of these proteins are encoded by
saprophytic and pathogenic fungi. In keeping with the fact
that fungi generally lack the Tat system, the fungal homologues have typical eukaryotic signal peptides, rather than
the longer and less hydrophobic Tat signal peptides. This is
suggestive that the gene for one of these proteins was
acquired by an ancestoral actinomycete through horizontal
transfer from a fungus and that a Tat-targeting sequence
was acquired to allow secretion of the protein in the
prokaryote. It is, however, not readily apparent why these
proteins should be substrates for the Tat pathway rather
than the Sec system – for example none of them are
predicted to bind redox cofactors that would necessitate
export in a folded conformation. Substrates for the Sec
machinery are transported in an unfolded form and usually
interact with cytoplasmic chaperones to maintain them in
an unfolded, export competent state. It is possible that
proteins coded by genes acquired by horizontal transfer
are not recognized by host chaperones as they are nonnative, and it therefore may be advantageous to export
such proteins in a folded form.
Of the seven Tat-secreted proteins affecting virulence,
SCAB06471 is homologous to an alpha-L-fucosidase
from G. zeae, the cause of wheat head blight and M.
oryzae, the rice blast pathogen (Oh et al., 2008). This
enzyme hydrolyses the alpha-1,6-linked fucose joined to
the reducing-end N-acetylglucosamine of carbohydrate
moieties in glycoproteins. SCAB80581 encodes a protein
for amine oxidase, both are homologous to proteins in
the nonpathogenic Streptomyces pristinaespiralis and
pathogenic Mycobacterium spp. SCAB81041, a putative
ABC transporter, is a periplasmic-binding protein and
homologous to proteins in Streptomyces sviceus, and in
Thermomicrobium roseum, a Gram negative, obligately
thermophilic bacterium.
Seven putative effector proteins chosen based on bioinformatic analysis did not have a virulence phenotype.
One of these was SCAB08951, which was highly similar
(Table 3) to proteins encoded by the Gram negative soil
bacterium Acidovorax delafieldii, the Gram negative plant
pathogen Dickeya dadantii (which both have predicted
Sec signal peptides) and the plant pathogenic oomycete
Phytophthora infestans (which has a predicted eukaryotic
signal peptide). Since effector proteins commonly have a
role in host specificity (Lindeberg et al., 2009), evaluation
of these mutants on additional host plants, particularly
potato, would be necessary.
Experimental procedures
Strains, plasmids, media and culture conditions
Streptomyces scabies strain 87-22 and the cognate DtatC
strain (see below) were routinely grown on either IPM (which
contained per litre of tap water 50 g SmashR instant potato
mash and 12 g agar), International Streptomyces Project
medium 2, or International Streptomyces Project medium 4
(BD Biosciences, San Jose, CA) at 28°C. For proteomic
analysis, strains were cultured on either IPM, SFM (Hobbs
et al., 1989), R5 medium (Thompson et al., 1980) or oat bran
broth medium (Goyer et al., 1998). For growth in liquid culture
strains were cultured aerobically in TSB (Kieser et al., 2000).
Phenotypic growth tests were carried out on DNA medium
(Kieser et al., 2000) and halo diameters were determined
using the image processing software GIMP (GNU Image
Manipulation Program – http://www.gimp.org/).
To test candidate signal peptides for Tat dependence in the
agarase assay, DNA encoding the signal peptides of interest
were cloned in-frame (as NdeI-BamHI/BglII fragments) with
the mature agarase sequence in the integrative plasmid
pTDW46H (which is identical to pTDW46 except that the
apramycin resistance specified by the vector has been
replaced by hygromycin resistance from pIJ10700 using the
REDIRECT method (Gust et al., 2003) and the oligonucleotide primers hyg_fwd and hyg_rev. The oligonucleotide
primer sequences used for signal peptide amplification are
listed in Table S4, and the plasmids used in this study are
listed in Table S5.
Agarase assays performed and quantified as described by
Widdick et al. (2006; 2008).
Construction and complementation of the S. scabies
DtatC strain
For construction of the S. scabies DtatC strain, SM1, an
approximately 1000 bp region directly upstream of the tatC
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
268 M. V. Joshi et al. 䊏
gene (SCAB73601) was amplified by polymerase chain reaction (PCR) using the oligonucleotides utatcscabf1 and utatcscabr (see Table S4 for a list of oligonucleotide primers used in
this study) and S. scabies chromosomal DNA as template,
digested with EcoRI and HindIII and cloned into pBluescript
(KS+) that had been similarly digested. An approximately
1000 bp region downstream of tatC was subsequently amplified using the oligonucleotides dtatcscabf and dtatcscabr1,
digested with EcoRI and XbaI and cloned into pBluescript
already containing the tatC upstream region that had been
similarly digested. The apramycin resistance cassette from
pIJ773 (Gust et al., 2003) was amplified using oligonucleotides ecoriapraf and ecoriaprar, cleaved with EcoRI and
cloned between the upstream and downstream regions of the
tatC deletion allele assembled in pBluescript. To allow the
detection of double-crossover strains by replica plating the
ampicillin/carbenicillin resistance cassette of the carrier
plasmid was replaced with a kanamycin resistance cassette,
which, unlike bla, is suitable for selection in Streptomyces.
This was achieved using the REDIRECT method of Gust et al.
(2003), and the kanamycin resistance cassette was amplified
by PCR using the oligonucleotides KanRtrans2_fwd and
KanRtrans2_rev (Table S4). The plasmid was transferred by
mating into S. scabies 87-22 and single-crossover recombinants were selected for on MS medium containing apramycin
and kanamycin (Gust et al., 2003). Double-crossover recombinants were subsequently selected by several rounds of
growth on non-selective media followed by selection for colonies that were apramycin-resistant and kanamycin-sensitive.
Loss of the tatC gene in strain SM1 was subsequently confirmed by PCR and by Southern blot analysis.
To test for complementation of the S. scabies DtatC strain,
a synthetic construct covering the S. scabies tatAC genes
(covering from 260 bp upstream of tatA to 60 bp downstream
of tatC) in pUC57 was purchased from GenScript, NJ, USA.
The tatAC-containing region was excised by digestion with
XbaI and cloned into similarly digested pSET-SOR-hyg
(Sean O’Rourke, unpublished) to give plasmid pTDW185,
which integrates site specifically into the Streptomyces
chromosome.
Deletion of genes encoding putative Tat substrates
Deletion mutants for 14 putative Tat-secreted virulence proteins (Tables 3 and S1) were created using the PCR-based
ReDirect gene disruption protocol (Gust et al., 2003). In this
case, a cosmid library of S. scabies strain 87-22 containing a
kanamycin resistance marker (neo) was used. Cosmid
clones containing the target genes were individually introduced into E. coli BW25113 carrying an arabinose-inducible
lRed-expressing plasmid pKD46 (AmpR) (Datsenko and
Wanner, 2000). A gene replacement cassette [aac(3)IV
(ApraR) + oriTRK2] from the plasmid pIJ773 (Gust et al.,
2003) was PCR-amplified using gene-specific redirect
primers (Table S4). The resulting PCR products were transformed into the cosmid-containing E. coli strain and selected
for apramycin resistance. The mutant cosmids were then
introduced into S. scabies 87-22 by intergeneric conjugation
from E. coli ET12567 (carrying the helper plasmid pUZ8002).
Exconjugants were screened for apramycin resistance
(100 mg ml-1) and kanamycin sensitivity (50 mg ml-1), indicat-
ing a double-crossover allelic exchange in S. scabies; deletions were confirmed by PCR analysis.
Protein methods
Growth curves for Streptomyces strains were monitored by
assaying total protein at regular intervals. To this end, 1 ¥ 106
spores of each strain were inoculated into 100 ml of TSB
medium and the cultures were incubated with shaking at
30°C over 56 h. Every 2 h, 3 ¥ 1 ml samples were withdrawn,
the cells pelleted and frozen at -20°C. The frozen samples
were subsequently resuspended in 1 ml 1N NaOH, 0.1%
SDS and boiled for 10 min to lyse the cells. The samples
were then clarified by centrifugation and 500 ml of the resulting supernatants were diluted with an equal volume of sterile
distilled water, and 25 ml of each sample was used for protein
determination using the Biorad DC Protein Assay Kit.
For the preparation of extracellular proteins, autoclaved
75 mm cellophane discs were placed onto solid media plates
(SFM, OBM, IPM or R5), inoculated with 106 spores of the S.
scabies WT and DtatC strains and incubated for 48 h at 30°C.
The biomass was scraped from the cellophane discs and
extracellular proteins were released following cell wall
washing exactly as described by Widdick et al., (2006). Separation of TCA-precipitated cell wall washes of S. scabies
extracellular proteins by two-dimensional gel electrophoresis
(2D-PAGE) was performed using the immobilized pH gradients in the pH range 3–10 as described (Antelmann et al.,
2001). For quantification of the relative protein amounts of
extracellular proteins that are decreased in the DtatC mutant
proteome compared with the WT, 200 mg protein were separated by 2D-PAGE and the resulting 2D gels were stained
with Coomassie-Brilliant Blue as described (Antelmann et al.,
2001).
Quantitative image analysis was performed from the
coomassie-stained 2D gels using the DECODON Delta 2D
software (http://www.decodon.com). The 2D gel images from
WT and the DtatC mutant cell wall proteomes were aligned
using a warp transformation. Before spot detection and quantification was performed, a fused 2D gel of both images was
created using the ‘union fusion’ algorithm of Delta2D. Spot
detection was performed in the fusion gel containing all spots
present in both images according to the automatically suggested parameters for background subtraction, average spot
size, and spot sensitivity. The resulting spot shapes were
reviewed and manually edited in the fusion gel if necessary.
This reviewed spot mask served as a spot detection consensus for all gel images, which was applied to both images to
guide the spot detection and quantification. This enables spot
quantification in all gels at the same locations resulting in
100% matching and in a reliable analysis of complete expression profiles. Normalization was performed by calculating the
quantity of each single spot in percentage related to the total
spot quantity per gel. Proteins showing an induction of at
least twofold in the WT compared with the DtatC mutant strain
are listed in Tables 1, S2 and S3.
For standard identification of the proteins from 2D gels,
spot cutting, tryptic digestion of the proteins and spotting of
the resulting peptides onto the MALDI-targets (Voyager
DE-STR, PerSeptive Biosystems) were performed using the
Ettan Spot Handling Workstation (Amersham-Biosciences,
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 269
Uppsala, Sweden) as described previously (Eymann et al.,
2004). The MALDI-TOF-TOF measurement of spotted
peptide solutions was carried out on a Proteome-Analyzer
4800 (Applied Biosystems, Foster City, CA, USA) as
described previously (Eymann et al., 2004). The mascot
search was performed against the available S. scabies database (http://www.sanger.ac.uk/Projects/S_scabies/).
Thaxtomin A extraction and quantification
Streptomyces scabies WT 87-22 and the DtatC mutant were
grown in 6 ¥ 5 ml of oat bran broth medium (Johnson et al.,
2007) in 6-well plates for 7 days at 25 ⫾ 2°C with moderate
shaking (~120 r.p.m.). Mycelia were pelleted by centrifugation and discarded. Thaxtomin A was extracted from culture
supernatants and was analysed by HPLC as previously
described (Johnson et al., 2007).
Plant virulence assays
Arabidopsis thaliana (ecotype Columbia) surface-sterilized
seeds were placed on MS) (Murashige and Skoog, 1962)
agar medium with 2% sucrose. Plants were grown at
21 ⫾ 2°C with a 16 h photoperiod for 7 days, then inoculated with S. scabies 87-22 (WT) or the cognate deletion
mutants (DtatC mutant or one of the 14 strains harbouring
deletions in genes encoding Tat substrates). In all cases,
spore suspensions (1x106 cfu) were applied to seedling root
tips. Disease symptoms were noted every week and images
were taken 21–28 days post inoculation. Each experiment
was repeated three times with five replicates of 20–25
plants.
Root colonization by the DtatC mutant strain
Streptomyces scabies WT 87-22 and the DtatC mutant
strains each tagged with a gene-encoding EGFP were
created and used for colonization studies with A. thaliana.
Vector pIJ8641 (Sun et al., 1999) or pRFSRL16 (R.F. Seipke,
unpublished), carrying the egfp gene downstream of the constitutive ermEp* promoter and an antibiotic-resistant marker
were replicated in the methylation-deficient E. coli strain
ET12567 prior to conjugation into S. scabies WT 87-22 and
the DtatC mutant strain respectively. A. thaliana seedlings
were grown on liquid MS medium with 2% sucrose for 7 days
and inoculated at the root tip with a spore suspension
(1 x 106 cfu). Laser scanning confocal microscopy was used
to visualize internal and external colonization of Arabidopsis
roots at 24 h intervals. The roots of harvested plants were
mounted in water immediately after harvesting and observed
using a Leica TCS SP5 confocal microscope (Leica Microsystems, Wetzlar, Germany), with a 10 ¥ and 63 ¥ water
immersion objectives as described (Joshi et al., 2007a).
EGFP was visualized using a 4-line argon laser with an
excitation wavelength of 488 nm and an emission wavelength
of 500 to 550 nm. Differential interference contrast images
were collected simultaneously with the fluorescence images
using the transmitted light detector and processed using
Leica LAS-AF software (version 1.8.2).
Acknowledgements
We thank R. Morosoli for providing us with the S. lividans tatC
strain. We also thank Kent Loeffler for photography of Arabidopsis plants and R.F. Seipke for providing pRFSRL16. We
further thank the Decodon company for support with the Delta
2D software. This work was supported by the CEU project
LHSG-CT-2004-005257, the BBSRC through Grant
BB/F009224/1 and the MRC via a Senior Non-Clinical Fellowship award to TP. SGM was supported by a PhD studentship
funded by the BBSRC and JKF by a joint SCRI/University of
Dundee PhD studentship. JKF acknowledges the Society for
General Microbiology for funding a research trip to Cornell.
Partial support was also provided by the National Research
Initiative of the United States Department of Agriculture Cooperative State Research, Education, and Extension Service,
Grant number 2008-35319-19202. We thank PCIC, Boyce
Thompson Institute (funding sources NSF DBI-0618969 and
Triad Foundation) for imaging facilities.
References
Antelmann, H., Tjalsma, H., Voigt, B., Ohlmeier, S., Bron, S.,
van Dijl, J.M., and Hecker, M. (2001) A proteomic view on
genome-based signal peptide predictions. Genome Res
11: 1484–1502.
Bendtsen, J.D., Nielsen, H., Widdick, D., Palmer, T., and
Brunak, S. (2005) Prediction of twin-arginine signal
peptides. BMC Bioinformatics 6: 167.
Berks, B.C. (1996) A common export pathway for proteins
binding complex redox cofactors? Mol Microbiol 22: 393–
404.
Bignell, D.R., Seipke, R.F., Huguet-Tapia, J.C., Chambers,
A.H., Parry, R.J., and Loria, R. (2010) Streptomyces
scabies 87-22 contains a coronafacic acid-like biosynthetic
cluster that contributes to plant-microbe interactions. Mol
Plant Microbe Interact 23: 161–175.
Birkó, Z., Sümegi, A., Vinnai, A., van Wezel, G., Szeszák, F.,
Vitális, S., et al. (1999) Characterization of the gene for
factor C, an extracellular signal protein involved in morphological differentiation of Streptomyces griseus. Microbiology 145: 2245–2253.
Bischoff, V., Cookson, S.J., Wu, S., and Scheible, W.R.
(2009) Thaxtomin A affects CESA-complex density,
expression of cell wall genes, cell wall composition, and
causes ectopic lignification in Arabidopsis thaliana
seedlings. J Exp Bot 60: 955–965.
Bogsch, E., Sargent, F., Stanley, N.R., Berks, B.C., Robinson,
C., and Palmer, T. (1998) An essential component of a novel
bacterial protein export system with homologues in plastids
and mitochondria. J Biol Chem 273: 18003–18006.
Bukhalid, R.A., and Loria, R. (1997) Cloning and expression
of a gene from Streptomyces scabies encoding a putative
pathogenicity factor. J Bacteriol 179: 7776–7783.
Büttner, D., and He, S.Y. (2009) Type III protein secretion in
plant pathogenic bacteria. Plant Physiol 150: 1656–1664.
Caldelari, I., Mann, S., Crooks, C., and Palmer, T. (2006) The
Tat pathway of the plant pathogen Pseudomonas syringae
is required for optimal virulence. Mol Plant Microbe Interact
19: 200–212.
Chater, K.F., Biró, S., Lee, K.J., Palmer, T., and Schrempf, H.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
270 M. V. Joshi et al. 䊏
(2010) The evolution of Streptomyces has involved an
amazingly complex extracellular biology. FEMS Microbiol
Rev 34: 171–198.
Chen, L.Y., Leu, W.M., Wang, K.T., and Lee, Y.H. (1992)
Copper transfer and activation of the Streptomyces apotyrosinase are mediated through a complex formation
between apotyrosinase and its trans-activator MelC1.
J Biol Chem 267: 20100–20107.
Christie, P.J., Atmakuri, K., Krishnamoorthy, V., Jakubowski,
S., and Cascales, E. (2005) Biogenesis, architecture, and
function of bacterial type IV secretion systems. Annu Rev
Microbiol 59: 4514–4585.
Cunnac, S., Lindeberg, M., and Collmer, A. (2009)
Pseudomonas syringae type III secretion system effectors:
repertoires in search of functions. Curr Opin Microbiol 12:
53–60.
Datsenko, K.A., and Wanner, B.L. (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using
PCR products. Proc Natl Acad Sci USA 97: 6640–6645.
De Buck, E., Lammertyn, E., and Anne, J. (2008) The importance of the twin-arginine translocation pathway for bacterial virulence. Trends Microbiol 16: 442–453.
Dilks, K., Rose, R.W., Hartmann, E., and Pohlschröder, M.
(2003) Prokaryotic utilization of the twin-arginine translocation pathway: a genomic survey. J Bacteriol 185: 1478–
1483.
Dilks, K., Giménez, M.I., and Pohlschröder, M. (2005)
Genetic and biochemical analysis of the twin-arginine
translocation pathway in halophilic archaea. J Bacteriol
187: 8104–8113.
Ding, Z., and Christie, P.J. (2003) Agrobacterium tumefaciens
twin-arginine-dependent translocation is important for virulence, flagellation, and chemotaxis but not type IV
secretion. J Bacteriol 185: 760–771.
Eymann, C., Dreisbach, A., Albrecht, D., Bernhardt, J.,
Becher, D., Gentner, S., et al. (2004) A comprehensive
proteome map of growing Bacillus subtilis cells. Proteomics 4: 2849–2876.
Flärdh, K., and Buttner, M.J. (2009) Streptomyces morphogenetics: dissecting differentiation in a filamentous
bacterium. Nat Rev Microbiol 7: 36–49.
Galán, J.E. (2009) Common themes in the design and
function of bacterial effectors. Cell Host Microbe 5: 571–
579.
Goyer, C., Vachon, J., and Beaulieu, C. (1998) Pathogenicity
of Streptomyces scabies Mutants Altered in Thaxtomin A
Production. Phytopathology 88: 44244–44245.
Gust, B., Challis, G.L., Fowler, K., Kieser, T., and Chater, K.F.
(2003) PCR-targeted Streptomyces gene replacement
identifies a protein domain needed for biosynthesis of the
sesquiterpene soil odor geosmin. Proc Natl Acad Sci USA
100: 1541–1546.
Healy, F.G., Wach, M., Krasnoff, S.B., Gibson, D.M., and
Loria, R. (2000) The txtAB genes of the plant pathogen
Streptomyces acidiscabies encode a peptide synthetase
required for phytotoxin thaxtomin A production and
pathogenicity. Mol Microbiol 38: 794–804.
Hicks, M.G., Guymer, D., Buchanan, G., Widdick, D.A.,
Caldelari, I., Berks, B.C., and Palmer, T. (2006) Formation
of functional Tat translocases from heterologous
components. BMC Microbiol 6: 64.
Hobbs, G., Frazer, C.M., Gardner, D.C.J., Cullum, J.A., and
Oliver, S.G. (1989) Dispersed growth of Streptomyces in
liquid culture. Appl Microbiol Biotechnol 31: 272–277.
Hogenhout, S.A., Van der Hoorn, R.A., Terauchi, R., and
Kamoun, S. (2009) Emerging concepts in effector biology
of plant-associated organisms. Mol Plant Microbe Interact
22: 115–122.
Ize, B., Stanley, N.R., Buchanan, G., and Palmer, T. (2003)
Role of the Escherichia coli Tat pathway in outer membrane
integrity. Mol Microbiol 48: 1183–1193.
Johnson, E.G., Joshi, M.V., Gibson, D.M., and Loria, R.
(2007) Cello-oligosaccarides released from host plants
induce pathogenicity in scab-causing Streptomyces
species. Physiol Mol Plant Path 71: 18–25.
Jongbloed, J.D., Grieger, U., Antelmann, H., Hecker, M.,
Nijland, R., Bron, S., and van Dijl, J.M. (2004) Two minimal
Tat translocases in Bacillus. Mol Microbiol 54: 1319–1325.
Joshi, M., Rong, X., Moll, S., Kers, J., Franco, C., and Loria,
R. (2007b) Streptomyces turgidiscabies secretes a novel
virulence protein, Nec1, which facilitates infection. Mol
Plant Microbe Interact 20: 599–608.
Joshi, M.V., Bignell, D.R., Johnson, E.G., Sparks, J.P.,
Gibson, D.M., and Loria, R. (2007a) The AraC/XylS
regulator TxtR modulates thaxtomin biosynthesis and
virulence in Streptomyces scabies. Mol Microbiol 66:
633–642.
Kers, J.A., Cameron, K.D., Joshi, M.V., Bukhalid, R.A.,
Morello, J.E., Wach, M.J., et al. (2005) A large, mobile
pathogenicity island confers plant pathogenicity on Streptomyces species. Mol Microbiol 55: 1025–1033.
Kieser, T., Bibb, M.J., Buttner, M.J., Chater, K.F., and
Hopwood, D.A. (2000) Practical Streptomyces Genetics.
Norfolk: The John Innes Foundation.
King, R.R., Lawrence, C.H., Clark, M.C., and Calhoun, L.A.
(1989) Isolation and characterization of phytotoxins associated with Streptomyces scabies. J Chem Soc Chem
Commun 13: 849–850.
Lee, P.A., Tullman-Ercek, D., and Georgiou, G. (2006) The
bacterial twin-arginine translocation pathway. Annu Rev
Microbiol 60: 373–395.
Leu, W.M., Chen, L.Y., Liaw, L.L., and Lee, Y.H. (1992)
Secretion of the Streptomyces tyrosinase is mediated
through its trans-activator protein, MelC1. J Biol Chem
267: 20108–20113.
Li, H., Jacques, P.E., Ghinet, M.G., Brzezinski, R., and Morosoli, R. (2005) Determining the functionality of putative
Tat-dependent signal peptides in Streptomyces coelicolor
A3(2) by using two different reporter proteins. Microbiology
151: 2189–2198.
Lindeberg, M., Cunnac, S., and Collmer, A. (2009) The evolution of Pseudomonas syringae host specificity and type III
effector repertoires. Mol Plant Pathol 10: 767–775.
Loria, R., Kers, J., and Joshi, M. (2006) Evolution of plant
pathogenicity in Streptomyces. Annu Rev Phytopathol 44:
469–487.
Loria, R., Bignell, D.R., Moll, S., Huguet-Tapia, J.C., Joshi,
M.V., Johnson, E.G., et al. (2008) Thaxtomin biosynthesis:
the path to plant pathogenicity in the genus Streptomyces.
Antonie Van Leeuwenhoek 94: 3–10.
Lovell, H.C., Mansfield, J.W., Godfrey, S.A., Jackson, R.W.,
Hancock, J.T., and Arnold, D.L. (2009) Bacterial evolution
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271
The Tat pathway in S. scabies 271
by genomic island transfer occurs via DNA transformation
in planta. Curr Biol 19: 158615–158690.
Miguélez, E.M., Hardisson, C., and Manzanal, M.B. (1999)
Hyphal death during colony development in Streptomyces
antibioticus: morphological evidence for the existence of a
process of cell deletion in a multicellular prokaryote. J Cell
Biol 145: 515–525.
Morrissey, J.P., Wubben, J.P., and Osbourn, A.E. (2000)
Stagonospora avenae secretes multiple enzymes that
hydrolyze oat leaf saponins. Mol Plant Microbe Interact 13:
1041–1052.
Murashige, T., and Skoog, F. (1962) A revised medium for
rapid growth and bioassays with tobacco tissue cultures.
Physiol Plant 15: 473–497.
Ochsner, U.A., Snyder, A., Vasil, A.I., and Vasil, M.L. (2002)
Effects of the twin-arginine translocase on secretion of
virulence factors, stress response, and pathogenesis. Proc
Natl Acad Sci USA 99: 8312–8317.
Oh, Y., Donofrio, N., Pan, H., Coughlan, S., Brown, D.E.,
Meng, S., et al. (2008) Transcriptome analysis reveals new
insight into appressorium formation and function in the rice
blast fungus Magnaporthe oryzae. Genome Biol 9: R85.
Palmer, T., Sargent, F., and Berks, B.C. (2005) Export of
complex cofactor-containing proteins by the bacterial Tat
pathway. Trends Microbiol 13: 175–180.
Pradel, N., Ye, C., Livrelli, V., Xu, J., Joly, B., and Wu, L.F.
(2003) Contribution of the twin arginine translocation
system to the virulence of enterohemorrhagic Escherichia
coli O157:H7. Infect Immun 71: 4908–4916.
Rigel, N.W., and Braunstein, M. (2008) A new twist on an old
pathway – accessory Sec systems. Mol Microbiol 69: 291–
302.
Rose, R.W., Brüser, T., Kissinger, J.C., and Pohlschröder, M.
(2002) Adaptation of protein secretion to extremely highsalt conditions by extensive use of the twin-arginine translocation pathway. Mol Microbiol 45: 943–950.
Saint-Joanis, B., Demangel, C., Jackson, M., Brodin, P., Marsollier, L., Boshoff, H., and Cole, S.T. (2006) Inactivation of
Rv2525c, a substrate of the twin arginine translocation
(Tat) system of Mycobacterium tuberculosis, increases
beta-lactam susceptibility and virulence. J Bacteriol 188:
6669–6679.
Sargent, F., Bogsch, E., Stanley, N.R., Wexler, M., Robinson,
C., Berks, B.C., and Palmer, T. (1998) Overlapping functions of components of a bacterial Sec-independent protein
export pathway. EMBO J 17: 3640–3650.
Sargent, F., Stanley, N.R., Berks, B.C., and Palmer, T. (1999)
Sec-independent protein translocation in Escherichia coli:
a distinct and pivotal role for the TatB protein. J Biol Chem
274: 36073–36083.
Schaerlaekens, K., Schierová, M., Lammertyn, E., Geukens,
N., Anné, J., and Van Mellaert, L. (2001) Twin-arginine
translocation pathway in Streptomyces lividans. J Bacteriol
183: 6727–6732.
Schaerlaekens, K., Van Mellaert, L., Lammertyn, E.,
Geukens, N., and Anné, J. (2004) The importance of the
Tat-dependent protein secretion pathway in Streptomyces
as revealed by phenotypic changes in tat deletion
mutants and genome analysis. Microbiology 150: 21–
31.
Scheible, W.R., Fry, B., Kochevenko, A., Schindelasch, D.,
Zimmerli, L., Somerville, S., et al. (2003) An Arabidopsis
mutant resistant to thaxtomin A, a cellulose synthesis
inhibitor from Streptomyces species. Plant Cell 15: 1781–
1794.
Seipke, R.F., and Loria, R. (2008) Streptomyces scabies
87-22 possesses a functional tomatinase. J Bacteriol 190:
7684–7692.
Shames, S.R., Auweter, S.D., and Finlay, B.B. (2009)
Co-evolution and exploitation of host cell signaling pathways by bacterial pathogens. Int J Biochem Cell Biol 41:
380–389.
Spallek, T., Robatzek, S., and Göhre, V. (2009) How
microbes utilize host ubiquitination. Cell Microbiol 11:
1425–1434.
Stanley, N.R., Palmer, T., and Berks, B.C. (2000) The twin
arginine consensus motif of Tat signal peptides is involved
in Sec-independent protein targeting in Escherichia coli.
J Biol Chem 275: 11591–11596.
Sun, J., Kelemen, G.H., Fernandez-Abalos, J.M., and Bibb,
M.J. (1999) Green fluorescent protein as a reporter for
spatial and temporal gene expression in Streptomyces
coelicolor A3(2). Microbiology 145: 2221–2227.
Thompson, C.J., Ward, J.M., and Hopwood, D.A. (1980) DNA
cloning in Streptomyces: resistance genes from antibioticproducing species. Nature 286: 525–527.
Weiner, J.H., Bilous, P.T., Shaw, G.M., Lubitz, S.P., Frost, L.,
Thomas, G.H., et al. (1998) A novel and ubiquitous system
for membrane targeting and secretion of cofactorcontaining proteins. Cell 93: 93–101.
Widdick, D.A., Dilks, K., Chandra, G., Bottrill, A., Naldrett, M.,
Pohlschroder, M., and Palmer, T. (2006) The twin-arginine
translocation pathway is a major route of protein export in
Streptomyces coelicolor. Proc Natl Acad Sci USA 103:
17927–17932.
Widdick, D.A., Eijlander, R.T., van Dijl, J.-M., Kuipers, O.,
and Palmer, T. (2008) A facile reporter system for the
experimental identification of twin arginine (Tat) signal
peptides from all Kingdoms of life. J Mol Biol 375: 595–
603.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 252–271